Aligning point cloud views using persistent feature histograms

By Radu Bogdan Rusu, Nico Blodow, Zoltán-Csaba Márton, Michael Beetz. 2008

1,084 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, Environmental Engineering, Geology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

In Environmental Engineering

In Geology

Citations in the corpus

PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation (2016, background): “Point features often encode certain statistical properties of points and are designed to be invariant to certain transformations, which are typically classified as intrinsic [2, 21, 3] or extrinsic [18, 17, 13, 10, 5].”
Dynamic Graph CNN for Learning on Point Clouds (2018, methodology): “ation and segmentation, two model tasks in the point cloud processing world. Traditional methods for solving these problems employ handcrafted features to capture geometric properties of point clouds [26, 38, 39]. More recently, the success of deep neural networks for image processing has motivated ”
Fast Point Feature Histograms (FPFH) for 3D registration (2009, methodology): “The Greedy Initial Alignment method described in our previous work [2] is very robust since it uses point cloud intrinsic, rotation invariant features.”
Deep Global Registration (2020, background): “Traditional hand-crafted features commonly summarize pairwise or higher-order relationships in histograms [20, 37, 40, 36, 35].”
Enforcing Geometric Constraints of Virtual Normal for Depth Prediction (2019, background): “The surface normal is an important ‘local’ feature for many point-cloud based applications such as registration [37] and object detection [20, 16].”
Self-Supervised Deep Learning on Point Clouds by Reconstructing Space (2019, background): “Traditionally, systems for such tasks have relied on the approximate computation of geometric features such as faces, edges or corners [31, 11] and hand-crafted features encoding statistical properties [3, 27].”
A Comprehensive Performance Evaluation of 3D Local Feature Descriptors (2015, background): “3 and include SI (Johnson and Hebert 1999), 3DSC (Frome et al. 2004), LSP (Chen and Bhanu 2007a), THRIFT (Flint et al. 2008), PFH (Rusu et al. 2008), FPFH (Rusu et al. 2009), SHOT (Tombari et al. 2010b), USC (Tombari et al. 2010a), RoPS (Guo et al. 2013b), and TriSI (Guo et al. 2013c).”
Pointwise Rotation-Invariant Network with Adaptive Sampling and 3D Spherical Voxel Convolution (2018, methodology): “fferent rotations. 2. Related Work 2.1. Learning from Geometries The development of features from geometries could be retrospected to manual designed features, including Point Feature Histograms (PFH)[23], Fast Point Feature Histograms (FPFH)[22], Signature of Histogram Orientations (SHOT)[24], and ”
The Perfect Match: 3D Point Cloud Matching With Smoothed Densities (2018, background): “On the other hand, several approaches [29, 28, 2] resort to a LRF-free representation based on intrinsically invariant features (e.”
Form2Fit: Learning Shape Priors for Generalizable Assembly from Disassembly (2019, background): “correspondences is a fundamental task in vision and graphics, studied extensively in prior work via descriptors [13], [14], [15], [16], [17], [18].”
3D Object Recognition in Cluttered Scenes with Local Surface Features: A Survey (2014, background): “[118] proposed “point feature histograms” (PFH) to encode a local surface.”
Towards 3D Point cloud based object maps for household environments (2008, methodology): “We propose the use of a better system that combines several aspects of the geometry of a point’s neighborhood for feature estimation [39], which already showed promising results in our previous work [29,32,30], as a multi-value feature set for selecting point correspondences.”
SoftGroup for 3D Instance Segmentation on Point Clouds (2022, methodology): “To process point clouds, early methods [2,3,36,37] extract hand-crafted features based on statistical properties of points.”
3DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructions (2016, methodology): “For Johnson et al. (Spin-Images) [19] and Rusu et al. (Fast Point Feature Histograms) [28], we use the implementation provided in the Point Cloud Library (PCL).”
Deep Learning on Point Sets for 3 D Classification and Segmentation (2016, background): “Point features often encode certain statistical properties of points and are designed to be invariant to certain transformations, which are typically classified as intrinsic [2, 21, 3] or extrinsic [18, 17, 13, 10, 5].”
PREDATOR: Registration of 3D Point Clouds with Low Overlap (2020, background): “Local 3D feature descriptors: Early local descriptors for point clouds [19, 31, 30, 37, 36] aimed to characterise the local geometry by using hand-crafted features.”
Fully Convolutional Geometric Features (2019, methodology): “PFH [24] and FPFH [23] build an oriented histogram using pairwise geometric properties.”
I3DOL: Incremental 3D Object Learning without Catastrophic Forgetting (2020, background): “Although enormous shape descriptors (Pauwels et al. 2013; Al-Osaimi 2016; Buch, Kiforenko, and Kraft 2017; Rusu et al. 2008; Sun, Ovsjanikov, and Guibas 2009) are handcrafted by domain experts for point cloud representation, they are invariant to the specific transformations, and cannot generalize…”
Scan Context: Egocentric Spatial Descriptor for Place Recognition Within 3D Point Cloud Map (2018, background): “Local descriptors, such as PFH [4], SHOT [5], shape context [7], or spin image [6], ﬁrst ﬁnd a keypoint, separate nearby points into bins, and encode a pattern of surrounding bins into a histogram.”
RPM-Net: Robust Point Matching Using Learned Features (2020, background): “PFH [27] and FPFH [26] design rotation invariant descriptors to avoid the need for a LRF.”